Search papers, labs, and topics across Lattice.
UC San Diego
8
0
10
3
AI-native SQL queries expose critical gaps in model performance, with top proprietary models still struggling to achieve 70% execution accuracy.
Local Branch Routing enables language models to leverage contextual evidence for decision-making without the computational burden of full solution searches, leading to substantial improvements in reasoning accuracy.
ACTS achieves full-thinking performance with up to 40% fewer tokens, enabling precise control over reasoning efficiency and accuracy.
Masking stale observations can boost search agent accuracy, but only under specific conditions—too much masking can backfire dramatically.
Diffusion models, typically too slow for interactive music generation, can now jam in real-time on a laptop thanks to a clever caching trick and a new alignment method.
Reranking in recommender systems can be revolutionized by shifting from local indices to generating global identifiers, enhancing robustness and user satisfaction.
Rollout design in LLM reinforcement learning is more than just sampling trajectories – it's a modular pipeline you can optimize for reliability, coverage, and cost.
Achieve near 20-point accuracy gains in reasoning tasks by dynamically routing between latent and discrete reasoning spaces based on model confidence.
